Induced Earthquake Sequences Research Articles

Erratum to The Influence of a Transitional Stress Regime on the Source Characteristics of Induced Seismicity and Fault Activation: Evidence from the 30 November 2018 Fort St. John ML 4.5 Induced Earthquake Sequence

<i>Erratum to</i> The Influence of a Transitional Stress Regime on the Source Characteristics of Induced Seismicity and Fault Activation: Evidence from the 30 November 2018 Fort St. John ML 4.5 Induced Earthquake Sequence

Fluid‐Induced Aseismic Slip May Explain the Non‐Self‐Similar Source Scaling of the Induced Earthquake Sequence Near the Dallas‐Fort Worth Airport, Texas

AbstractNumerous studies have reported the occurrence of aseismic slip or slow slip events along faults induced by fluid injection. However, the underlying physical mechanism and its impact on induced seismicity remain unclear. In this study, we develop a numerical model that incorporates fluid injection on a fault governed by rate‐and‐state friction to simulate the coupled processes of pore‐pressure diffusion, aseismic slip, and dynamic rupture. We establish a field‐scale model to emulate the source characteristics of induced seismicity near the Dallas‐Fort Worth Airport (DFWA), Texas, where events with lower‐stress drops have been observed. Our numerical calculations reveal that the diffusion of fluid pressure modifies fault criticality and induces aseismic slip with lower stress drop values (&lt;1 MPa), which further influence the timing and source properties of subsequent seismic ruptures. We observe that the level of pore‐pressure perturbation exhibits a positive correlation with aseismic‐stress drops but a reversed trend with seismic‐stress drops. Simulations encompassing diverse injection operations and fault frictional parameters generate a wide spectrum of slip modes, with the scaling relationship of moment (M0) with ruptured radius (r0) following an unusual trend, , similar to observed in the DFWA sequence. Based on the consistent scaling, we hypothesize that the lower‐stress‐drop events in the DFWA may imply less dynamic ruptures in the transition from aseismic to seismic slip, located in the middle of the broad slip spectrum, as illustrated in our simulations.

Influence of Fault Architecture on Induced Earthquake Sequence Evolution Revealed by High‐Resolution Focal Mechanism Solutions

AbstractThe increasing seismicity and improved seismic observation network in recent years provide an opportunity to explore factors that influence the triggering processes, spatiotemporal evolution, and maximum magnitude of induced sequences. We map the fault architecture and stress state of four induced sequences in Oklahoma to determine their influence on the seismicity. We systematically relocate the earthquakes and compute hundreds of focal mechanisms of small to medium events (1.0 &lt; M &lt; 5.1) using various techniques, including machine learning, for the Guthrie, Woodward, Cushing, and Fairview sequences. The detailed fault geometry and spatiotemporal evolution of seismicity and stress states reveal different dominant driving forces for each sequence. In Cushing and Fairview (largest event ≥M5.0), the main fault structures are near‐vertical narrow strike‐slip faults, with most of the small earthquake fault planes optimally oriented. The two sequences exhibit discontinuous temporal migration but strong earthquake self‐driven rupture growth. In Guthrie and Woodward (largest event &lt;M5.0), the two sequences show more complex diffuse fault structures with varying dipping angles along depth. The inverted focal mechanisms show a mix of strike‐slip faulting and normal faulting in both sequences, and the normal faulting events are less optimally oriented than strike‐slip events. The two sequences are dominated by continuous diffusive migration in time driven by pore pressure propagation. The above results suggest that fault architecture and stress state influence sequence evolution, major driving forces, and possibly maximum magnitude.

The Influence of a Transitional Stress Regime on the Source Characteristics of Induced Seismicity and Fault Activation: Evidence from the 30 November 2018 Fort St. John ML 4.5 Induced Earthquake Sequence

ABSTRACTOn 30 November 2018, a sequence of seismicity including a felt (ML∼4.5) induced earthquake occurred ∼16 km southwest of Fort St. John, British Columbia. Using a local seismograph network around the epicentral region, we identified &amp;gt; 560 seismic events over a two-week period, incorporating two mainshock events within a 45 min time interval, both with ML&amp;gt;4.3. This seismicity occurred close in location and depth to ongoing hydraulic fracturing operations. Using previously unpublished data, our analysis suggests that events, including the largest mainshock, occurred at the interval of fluid injection, which is shallower than previously reported. The events showed a mix of reverse, oblique normal, and strike-slip mechanisms within a well-defined structural corridor that forms the southern margin of the Fort St. John graben. The two mainshock events reveal opposing mechanisms: one as a reverse (re)activation of a normal fault (ML 4.5) and the other an oblique normal mechanism (ML 4.3). Stress inversion and bootstrap analysis of 72 well-constrained focal mechanisms indicate that the maximum principal stress direction is horizontal, oriented in a north-northeast direction (3°–36°). However, the intermediate and minimum stress axes fluctuate between horizontal and vertical and are nearly equal in magnitude, indicating that both reverse and strike-slip regimes can occur in response to relatively small stress perturbations. Stress inversions using event subsets before and after the largest mainshock reveals an approximately 30° counter-clockwise coseismic rotation of the principal stress axes in the hypocentral region. Furthermore, the observed seismicity suggests that the largest mainshock event exceeded the calculated Mmax using models based on injected volumes, suggesting that it may be an example of runaway rupture. This has important implications for risk analysis, because small changes in the stress field may be induced through ongoing operations in this area, destabilizing different faults within a complex structural environment.

Correlation Between Poroelastic Stress Perturbation and Multidisposal Wells Induced Earthquake Sequence in Cushing, Oklahoma

AbstractOver 100 small‐ to moderate‐sized earthquakes, including an Mw 5.0 event, were detected during September 2015 to November 2016 near the town of Cushing, Oklahoma. The seismic sequence was spatial‐temporally linked to four wastewater disposal wells within 4 km. We calculate pore pressure and stress perturbations caused by fluid injection at multiple wells and analyze seismic risk in a Coulomb failure stress framework. Despite being more than an order of magnitude smaller than the pore pressure perturbation, the sign of shear stress change, in the sense of assumed right‐lateral fault motion, dictates where earthquakes are induced. Most of the relocated earthquakes are located within areas of positive shear stress changes. Our results suggest that poroelastic stress changes also play an essential role in the wastewater disposal environment, and a strategic design of well locations with respect to fault orientation and direction of motion can help mitigate induced seismic hazard.

Revisiting the Timpson Induced Earthquake Sequence: A System of Two Parallel Faults

AbstractThe 17 May 2012 M4.8 Timpson earthquake is the largest known earthquake in eastern Texas. It is thought to have been induced by wastewater injection from two nearby, high‐volume wells. Its cataloged aftershocks form a NW‐SE trend, which unlike other induced earthquakes sequences is unfavorably oriented for failure in the local stress field. To understand this, we enriched the catalog using PhaseNet, a deep‐learning‐based picker followed by double‐difference relocation with cross‐correlation‐based differential traveltimes. We clustered the aftershocks based on waveform similarity. Most of the seismicity falls into two clusters, which define a detailed fault structure of two parallel subfaults that are more favorably oriented than the overall trend. We inferred from waveform similarity that the sequence initiated on the northern subfault with a M3.9 foreshock and M4.8 mainshock, then extended to the southern subfault with a M4.1 aftershock, and was finally reactivated on the northern subfault with two more M4 events.

Improved approach for stress drop estimation and its application to an induced earthquake sequence in Oklahoma

SUMMARYWe calculate source parameters for fluid-injection induced earthquakes near Guthrie, Oklahoma, guided by synthetic tests to quantify uncertainties. The average stress drop during an earthquake is a parameter fundamental to ground motion prediction and earthquake source physics, but it has proved hard to measure accurately. This has limited our understanding of earthquake rupture, as well as the spatio-temporal variations of fault strength. We use synthetic tests based on a joint spectral-fitting method to define the resolution limit of the corner frequency as a function of the maximum frequency of usable signal, for both individual spectra and the average from multiple stations. Synthetic tests based on stacking analysis find that an improved stacking approach can recover the true input stress drop if the corner frequencies are within the resolution limit defined by joint spectral-fitting. We apply the improved approach to the Guthrie sequence, using different wave types and signal-to-noise criteria to understand the stability of the calculated stress drop values. The results suggest no systematic scaling relationship of stress drop for M ≤ 3.1 earthquakes, but larger events (M ≥ 3.5) tend to have higher average stress drops. Some robust spatio-temporal variations can be linked to the triggering processes and indicate possible stress heterogeneity within the fault zone. Tight clustering of low stress drop events at the beginning stage of the sequence suggests that pore pressure influences earthquake source processes. Events at shallow depth have lower stress drop compared to deeper events. The largest earthquake occurred within a cluster of high stress drop events, likely rupturing a strong asperity.

Activation of optimally and unfavourably oriented faults in a uniform local stress field during the 2011 Prague, Oklahoma, sequence

SUMMARYThe orientations of faults activated relative to the local principal stress directions can provide insights into the role of pore pressure changes in induced earthquake sequences. Here, we examine the 2011 M 5.7 Prague earthquake sequence that was induced by nearby wastewater disposal. We estimate the local principal compressive stress direction near the rupture as inferred from shear wave splitting measurements at spatial resolutions as small as 750 m. We find that the dominant azimuth observed is parallel to previous estimates of the regional compressive stress with some secondary azimuths oriented subparallel to the strike of the major fault structures. From an extended catalogue, we map ten distinct fault segments activated during the sequence that exhibit a wide array of orientations. We assess whether the five near-vertical fault planes are optimally oriented to fail in the determined stress field. We find that only two of the fault planes, including the M 5.7 main shock fault, are optimally oriented. Both the M 4.8 foreshock and M 4.8 aftershock occur on fault planes that deviate 20–29° from the optimal orientation for slip. Our results confirm that induced event sequences can occur on faults not optimally oriented for failure in the local stress field. The results suggest elevated pore fluid pressures likely induced failure along several of the faults activated in the 2011 Prague sequence.

Narrow Spatial Aftershock Zones for Induced Earthquake Sequences in Oklahoma

AbstractThe current paradigm for estimating regional long‐term seismic hazard involves declustering whereby aftershocks are removed from an earthquake catalog to identify the underlying background seismicity rate. Inaccurate declustering can ultimately underestimate or overestimate the regional seismic hazard. In Oklahoma, estimating a background seismicity rate is complicated by the highly variable seismicity rate over the last decade. To improve conventional declustering methods used in hazard modeling in Oklahoma, we scrutinize the aftershock windows used for declustering and investigate how aftershocks decay in space and time to establish data‐driven parameters for aftershock windowing. We observe that the spatial decay of aftershocks is more rapid in Oklahoma than in Southern California, motivating the need for smaller spatial declustering windows in Oklahoma. Temporal aftershock decay is statistically indistinguishable between Oklahoma and Southern California, suggesting that temporal declustering windows derived for Southern California are likely sufficient for Oklahoma.

A Shallow Earthquake Swarm Close to Hydrocarbon Activities: Discriminating between Natural and Induced Causes for the 2018–2019 Surrey, United Kingdom, Earthquake Sequence

ABSTRACTEarthquakes induced by subsurface industrial activities are a globally emotive issue, with a growing catalog of induced earthquake sequences. However, attempts at discriminating between natural and induced causes, particularly for anomalously shallow seismicity, can be challenging. An earthquake swarm during 2018–2019 in southeast England with a maximum magnitude of ML 3.2 received great public and media attention because of its proximity to operating oilfields. It is therefore vital and timely to provide a detailed characterization of the earthquake sequence at present, and to decide based on current evidence, whether the earthquakes were likely natural or induced. We detected 168 low‐magnitude earthquakes and computed detailed source parameters of these events. Most earthquakes occurred at a shallow depth of 2.3 km, &gt;1 km deeper than the geological formations targeted by the oilfields, and laterally &gt;3 km away from the drill sites. We combine the east–west‐trending cluster of the seismicity with 2D seismic reflection profiles to find the causative fault system for the earthquakes. A b‐value close to unity and strike‐slip faulting mechanisms are consistent with tectonic reactivation along a pre‐existing fault. Overall, we find no indicators in the earthquake parameters that would strongly suggest an induced source. Nor do we find any clear trends between seismicity and drilling activities based on operational logs provided by the operators. Injected volumes are near zero and monthly production amounts are many orders of magnitude smaller than other reported cases of extraction‐induced seismicity. On balance, and based on the available evidence, we find it currently unlikely that nearby industrial activities induced the seismic swarm. Most likely, the Surrey earthquakes offer a uniquely detailed insight into shallow seismicity within sedimentary basins. Nevertheless, self‐reporting of injection and production times and volumes by operators, and the lack of easily and publicly available oilfield operational data continues to be a point of concern for local residents.

Aftershock deficiency of induced earthquake sequences during rapid mitigation efforts in Oklahoma

Induced seismicity provides a rare opportunity to study earthquake triggering and underlying stress perturbations. Triggering can be a direct result of induced stress changes or indirect due to elastic stress transfer from preceding events leading to aftershocks. Both of these processes are observable in areas with larger magnitude induced events, such as Oklahoma. We study aftershock sequences of M2.5 to M5.8 earthquakes and examine the impact of targeted injection rate reductions. In comparing aftershock productivity between California and Oklahoma, we find similar exponential scaling statistics between mainshock magnitude and average number of aftershocks. For events with M≥4.5 Oklahoma exhibits several mainshocks with total number of aftershocks significantly below the average scaling behavior. The sequences with deficient aftershock numbers also experienced rapid, strong mitigation and reduced injection rates, whereas two events with M4.8 and M5.0 with weak mitigation exhibit normal aftershock productivity. The timing of when aftershock activity is reduced correlates with drops in injection rates with a lag time of several days. Large mainshocks with significantly reduced aftershocks may explain decreasing seismicity rates while seismic moment release was still increasing in Oklahoma in 2016. We investigate the expected poroelastic stress perturbations due to injection rate changes within a layered axisymmetric model and find that stresses are lowered by 10s to 100s kPa within the injection-affected zone. For earthquakes induced by poroelastic stress-increase at several kilometers from wells, the rapid shut-in of wells may lead to elastic stress reductions sufficiently high to arrest unfolding aftershock sequences within days after mitigation starts.

Hazard Implications of the 2016 Mw 5.0 Cushing, OK Earthquake from a Joint Analysis of Damage and InSAR Data

The Cushing Hub in Oklahoma, one of the largest oil storage facilities in the world, is federally designated as critical national infrastructure. In 2014, the formerly aseismic city of Cushing experienced a Mw 4.0 and 4.3 induced earthquake sequence due to wastewater injection. Since then, an M4+ earthquake sequence has occurred annually (October 2014, September 2015, November 2016). Thus far, damage to critical infrastructure has been minimal; however, a larger earthquake could pose significant risk to the Cushing Hub. In addition to inducing earthquakes, wastewater injection also threatens the Cushing Hub through gradual surface uplift. To characterize the impact of wastewater injection on critical infrastructure, we use Differential Interferometric Synthetic Aperture Radar (DInSAR), a satellite radar technique, to observe ground surface displacement in Cushing before and during the induced Mw 5.0 event. Here, we process interferograms of Single Look Complex (SLC) radar data from the European Space Agency (ESA) Sentinel-1A satellite. The preearthquake interferograms are used to create a time series of cumulative surface displacement, while the coseismic interferograms are used to invert for earthquake source characteristics. The time series of surface displacement reveals 4–5.5 cm of uplift across Cushing over 17 months. The coseismic interferogram inversion suggests that the 2016 Mw 5.0 earthquake is shallower than estimated from seismic inversions alone. This shallower source depth should be taken into account in future hazard assessments for regional infrastructure. In addition, monitoring of surface deformation near wastewater injection wells can be used to characterize the subsurface dynamics and implement measures to mitigate damage to critical installations.

The spatial footprint of injection wells in a global compilation of induced earthquake sequences.

Fluid injection can cause extensive earthquake activity, sometimes at unexpectedly large distances. Appropriately mitigating associated seismic hazards requires a better understanding of the zone of influence of injection. We analyze spatial seismicity decay in a global dataset of 18 induced cases with clear association between isolated wells and earthquakes. We distinguish two populations. The first is characterized by near-well seismicity density plateaus and abrupt decay, dominated by square-root space-time migration and pressure diffusion. Injection at these sites occurs within the crystalline basement. The second population exhibits larger spatial footprints and magnitudes, as well as a power law-like, steady spatial decay over more than 10 kilometers, potentially caused by poroelastic effects. Far-reaching spatial effects during injection may increase event magnitudes and seismic hazard beyond expectations based on purely pressure-driven seismicity.

Hydromechanical Earthquake Nucleation Model Forecasts Onset, Peak, and Falling Rates of Induced Seismicity in Oklahoma and Kansas

AbstractThe earthquake activity in Oklahoma and Kansas that began in 2008 reflects the most widespread instance of induced seismicity observed to date. We develop a reservoir model to calculate the hydrologic conditions associated with the activity of 902 saltwater disposal wells injecting into the Arbuckle aquifer. Estimates of basement fault stressing conditions inform a rate‐and‐state friction earthquake nucleation model to forecast the seismic response to injection. Our model replicates many salient features of the induced earthquake sequence, including the onset of seismicity, the timing of the peak seismicity rate, and the reduction in seismicity following decreased disposal activity. We present evidence for variable time lags between changes in injection and seismicity rates, consistent with the prediction from rate‐and‐state theory that seismicity rate transients occur over timescales inversely proportional to stressing rate. Given the efficacy of the hydromechanical model, as confirmed through a likelihood statistical test, the results of this study support broader integration of earthquake physics within seismic hazard analysis.

Temporal Correlation Between Seismic Moment and Injection Volume for an Induced Earthquake Sequence in Central Oklahoma

AbstractThe rapidly increased earthquake rate in the central United States has been linked with wastewater injection. While the overall understanding appears clear at large scales, the interaction between injection and faulting at smaller scales within individual sequences is still not clear. For an earthquake sequence in central Oklahoma, we conduct finer‐scale analysis of the spatiotemporal evolution of seismicity and pore pressure modeling. The pore pressure modeling suggests that nearby wells show much stronger correlation with earthquake sequence evolution. Detailed temporal analysis found correlation between earthquake rate, seismic moment, and injection rates from wells in close proximity. However, the observed maximum magnitude (Mmax) is about 1 order of magnitude smaller than expected based on a theoretical relationship between Mmax and cumulative volume. This discrepancy may point toward additional parameters, such as fault size and stress, which influence Mmax. The lower Mmax is consistent with the truncated Gutenberg‐Richter distribution observed from matched filter detected catalog. Overall, the detailed observations suggest that it is possible to resolve relationships between individual disposal wells and induced earthquake sequences.

HiQuake: The Human‐Induced Earthquake Database

ABSTRACT HiQuake —The Human‐Induced Earthquake Database is the most complete database of anthropogenic projects proposed, on scientific grounds, to have induced earthquake sequences. It is freely available to download from the website given in Data and Resources. At the time this article was written, HiQuake contained ∼730 anthropogenic projects proposed to have induced earthquakes, as well as associated project‐related and seismic data. The most commonly reported anthropogenic activities proposed to have induced earthquakes are mining and water reservoir impoundment. In recent years, the number of earthquake sequences proposed to have been induced by fluid‐injection activities has grown. The most commonly reported maximum observed magnitude in an induced earthquake sequence is 3≤ M MAX HiQuake proposed to have been induced had a magnitude of M w 7.9 and occurred in China. Such large earthquakes release mostly stress of natural tectonic origin, but are conceivably triggered by small anthropogenic stress changes. The data in HiQuake are of variable quality because they are drawn from publications that span almost a century. We estimate underreporting to be ∼30% for M ∼4 events, ∼60% for M ∼3 events, and ∼90% for M ∼2 events. The degree of certitude that the given earthquake sequences were anthropogenically induced is variable. HiQuake includes all earthquake sequences proposed on scientific grounds to have been human induced without regard to the strength of the case made. HiQuake is offered freely as a resource to interested parties, and judging the reliability of any particular case is the responsibility of the database user. HiQuake will be routinely updated to correct errors, update existing entries, and add new entries. It has the potential to help improve our understanding of induced earthquakes and to manage their impact on society.

The induced earthquake sequence related to the St. Gallen deep geothermal project (Switzerland): Fault reactivation and fluid interactions imaged by microseismicity

AbstractIn July 2013, a sequence of more than 340 earthquakes was induced by reservoir stimulations and well‐control procedures following a gas kick at a deep geothermal drilling project close to the city of St. Gallen, Switzerland. The sequence culminated in anML3.5 earthquake, which was felt within 10–15 km from the epicenter. High‐quality earthquake locations and 3‐D reflection seismic data acquired in the St. Gallen project provide a unique data set, which allows high‐resolution studies of earthquake triggering related to the injection of fluids into macroscopic fault zones. In this study, we present a high‐precision earthquake catalog of the induced sequence. Absolute locations are constrained by a coupled hypocenter‐velocity inversion, and subsequent double‐difference relocations image the geometry of theML3.5 rupture and resolve the spatiotemporal evolution of seismicity. A joint interpretation of earthquake and seismic data shows that the majority of the seismicity occurred in the pre‐Mesozoic basement, hundreds of meters below the borehole and the targeted Mesozoic sequence. We propose a hydraulic connectivity between the reactivated fault and the borehole, likely through faults mapped by seismic data. Despite the excellent quality of the seismic data, the association of seismicity with mapped faults remains ambiguous. In summary, our results document that the actual hydraulic properties of a fault system and hydraulic connections between its fault segments are complex and may not be predictable upfront. Incomplete knowledge of fault structures and stress heterogeneities within highly complex fault systems additionally challenge the degree of predictability of induced seismicity related to underground fluid injections.

Collective properties of injection‐induced earthquake sequences: 1. Model description and directivity bias

AbstractInduced seismicity is of increasing concern for oil and gas, geothermal, and carbon sequestration operations, with several M &gt; 5 events triggered in recent years. Modeling plays an important role in understanding the causes of this seismicity and in constraining seismic hazard. Here we study the collective properties of induced earthquake sequences and the physics underpinning them. In this first paper of a two‐part series, we focus on the directivity ratio, which quantifies whether fault rupture is dominated by one (unilateral) or two (bilateral) propagating fronts. In a second paper, we focus on the spatiotemporal and magnitude‐frequency distributions of induced seismicity. We develop a model that couples a fracture mechanics description of 1‐D fault rupture with fractal stress heterogeneity and the evolving pore pressure distribution around an injection well that triggers earthquakes. The extent of fault rupture is calculated from the equations of motion for two tips of an expanding crack centered at the earthquake hypocenter. Under tectonic loading conditions, our model exhibits a preference for unilateral rupture and a normal distribution of hypocenter locations, two features that are consistent with seismological observations. On the other hand, catalogs of induced events when injection occurs directly onto a fault exhibit a bias toward ruptures that propagate toward the injection well. This bias is due to relatively favorable conditions for rupture that exist within the high‐pressure plume. The strength of the directivity bias depends on a number of factors including the style of pressure buildup, the proximity of the fault to failure and event magnitude. For injection off a fault that triggers earthquakes, the modeled directivity bias is small and may be too weak for practical detection. For two hypothetical injection scenarios, we estimate the number of earthquake observations required to detect directivity bias.

The 2001-Present Induced Earthquake Sequence in the Raton Basin of Northern New Mexico and Southern Colorado

We investigate the ongoing seismicity in the Raton Basin and find that the deep injection of wastewater from the coal-bed methane field is responsible for inducing the majority of the seismicity since 2001. Many lines of evidence indicate that this earthquake sequence was induced by wastewater injection. First, there was a marked increase in seismicity shortly after major fluid injection began in the Raton Basin in 1999. From 1972 through July 2001, there was one M ≥ 4 earthquake in the Raton Basin, whereas 12 occurred between August 2001 and 2013. The statistical likelihood that such a rate change would occur if earthquakes behaved randomly in time is 3.0%. Moreover, this rate change is limited to the area of industrial activity. Earthquake rates remain low in the surrounding area. Second, the vast majority of the seismicity is within 5 km of active disposal wells and is shallow, ranging between 2 and 8 km depth. The two most carefully studied earthquake sequences in 2001 and 2011 have earthquakes within 2 km of high-volume, high-injection-rate wells. Third, injection wells in the area are commonly very high volume and high rate. Two wells adjacent to the August 2011 M 5.3 earthquake injected about 4.9 million cubic meters of wastewater before the earthquake, more than seven times the amount injected at the Rocky Mountain Arsenal well that caused damaging earthquakes near Denver, Colo- rado, in the 1960s. The August 2011 M 5.3 event is the second-largest earthquake to date for which there is clear evidence that the earthquake sequence was induced by fluid injection.

High-precision relocation and focal mechanism of the injection-induced seismicity at the Basel EGS

In early December 2006, a massive fluid injection was carried out at 5km depth below the city of Basel, Switzerland, for geothermal reservoir enhancement. During the six-day stimulation, approximately 13,000 induced microearthquakes were detected by a borehole network. The largest of the induced earthquakes, which had a magnitude of ML 3.4, was strongly felt in the Basel area and led to the termination of the project after only six days of stimulation. We analyzed the approximately 3500 locatable events of this induced earthquake sequence, which is one of the most densely monitored deep fluid-injections in the world. The seismic monitoring system consisted of six borehole seismometers at depths between 300m and 2700m near the injection well and of numerous surface stations in the Basel area. In this article, we report on the analysis of the sequence using exclusively data from the down-hole instruments. We show how a refinement of arrival time picks by cross-correlation techniques and subsequent high-precision relocations lead to significant improvements of the hypocenter locations compared to routinely applied manual procedures. We also analyze focal mechanisms determined from both first-motion polarities and amplitudes of signals recorded by the borehole sensors alone and compare the results to the focal mechanisms of the larger events recorded also by the surface networks. Our findings indicate that the induced sequence consists to a large part of repeating earthquakes that repeatedly rupture the same fault patches, and that the activated seismogenic structure is not a single fault plane but a heterogeneous fault zone.